Method and apparatus for obscuring features of an image

Abstract

A method and system that obscures image features in a region of interest designated by a user. An operator designates a special shaped region of interest and a computer convolves a specially designed kernel with the image data in the region. The kernel is designed to pull features outside the region into the region while combining the information within the region with the information pulled from outside. The kernel can be designed to correspond to the shape of the region for efficient computation. Once the region of interest is obscured the image is printed resulting in a print that does not include undesirable features.

What is claimed is: 
     
       1. A method of obscuring information in an image, comprising: (a) obtaining one digital representative of the image;   (b) completely enclosing a region of the image having a plurality of pixels for obscuration so that the lines delineate the enclosed region;   (c) merging first image signal information that is from outside the region and that is also in the digital representation with second image signal information that is from inside the region and that is also in the digital representation by applying at least two convolution kernels around the perimeter of the region toward the center in discrete sequential concentric smaller layers, the result of each layer being merged with the next layer; and,   (d) obscuring the bounded region as a result of step (c).   
     
     
       2. A system, comprising: an image source device providing one digital representative of an image;   an input/output device allowing a user to completely enclose a region of the image having a plurality of pixels for obscuration so that the lines delineate the enclosed region; and   a computer, coupled to said source device and said input/output device, processing the region of the image to obscure information in the region by merging signal information that is outside the region and that is also in the digital representation with signal information that is inside the region and that is also in the digital representation by applying at least two different convolution kernels around the perimeter of the region toward the center in discrete sequential concentric smaller layers, the result of each layer being merged with the next layer.   
     
     
       3. A method as recited in claim 1, wherein the region has a shape and the at least two convolution kernels has a shape corresponding to the shape of the region. 
     
     
       4. A method as recited in claim 1, wherein the at least two convolution kernels comprise an element for an image pixel of interest and elements closer to an outside edge of the region than the pixel of interest. 
     
     
       5. A method as recited in claim 4, wherein the at least two convolution kernels include elements that are a power of two. 
     
     
       6. A method as recited in claim 1, wherein the at least two convolution kernels comprise a first kernel for flat region areas and a second kernel for curved region areas. 
     
     
       7. A method as recited in claim 1, wherein the at least two convolution kernels are symmetric. 
     
     
       8. A method as recited in claim 7, wherein the at least two convolution kernels convolve a pixel of interest and exterior neighboring pixels. 
     
     
       9. A method as recited in claim 8, wherein the at least two convolution kernels convolve exterior neighbor pixels of an adjacent different layer. 
     
     
       10. A method as recited in claim 7, wherein the at least two convolution kernels are rotationally symmetric about a normal to a surface of the region. 
     
     
       11. A method as recited in claim 10, wherein the at least two convolution kernels provide the same response when symmetrically rotated when presented with identical data. 
     
     
       12. A method as recited in claim 9, wherein a weighting of the pixel of interest is 1/16th. 
     
     
       13. A method as recited in claim 9, wherein a weighting of the exterior immediate neighbor pixels is 4/16ths and a weighting of the exterior neighbors is 1/16th. 
     
     
       14. A method as recited in claim 1, wherein the at least two convolution kernels include a number of elements that are a power of two. 
     
     
       15. A method as recited in claim 1, wherein step (b) includes only addition and shift operations. 
     
     
       16. A method as recited in claim 1, wherein step (b) convolves information exterior to the region toward a central point of the region. 
     
     
       17. A method as recited in claim 1, wherein the region is a closed curve shaped to approximate an expected shape of a region of undesirable features. 
     
     
       18. A method as recited in claim 1, further comprising (c) printing the image. 
     
     
       19. A system as recited in claim 2, further comprising a printer printing the image including the obscured region. 
     
     
       20. A system according to claim 2 wherein the image source device is a camera or an image storage device.